The present invention relates to a sign for conveying information to people and machines, and more specifically, to a method and a device for automatically converting a sign and a method and a device for automatically reading a sign.
Nowadays, various signs are widely used to convey information to people in urban and rural areas. These signs may be texts such as a sign “P” that denotes a parking lot, graphics such as a sign “” that denotes a subway station, or a combination of texts and graphics. Since these signs are made using natural languages or graphics, people can easily read them and understand their meanings. However, sometimes it is not easy for machines to read these signs. Conventionally, a machine with a camera takes an image of a sign, and reads the sign and understands its meaning through image recognition. But for certain signs, especially those signs that are complicated or have a small size, it is hard to accurately read and understand them through the aforesaid conventional method, due to insufficient accuracy of the image recognition and difficulty in understanding the natural language per se.
In order to improve readability of a sign to a machine, many solutions have been provided. In one solution, information indicating locations and meanings of various signs placed in a city is collected and stored in a database, so that the machine can query the database to learn the meaning of the sign. However, such solution requires manual collection of the aforesaid information, which is troublesome. In another solution, a machine readable symbol (e.g., a bar code or the like) that denotes the meaning of the sign is printed together with the sign, so that the machine can understand the meaning of the sign by reading the symbol. However, the symbol employed by such solution will occupy space that could have been used for the sign, so that a size of the sign is decreased and thereby its readability is reduced.